Ext3 users - Apr 2003 - Identify file(s) on specific disk sector

When tar'ing up an ext3 filesystem on a 3ware RAID I have, I got
the following errors in my system log:

Apr  7 21:27:46 allo 3w-xxxx[957]: Drive error encountered on port 3 on
controller ID:0. Check cables and drives for media errors. (0xa)
Apr  7 21:27:46 allo kernel: 3w-xxxx: scsi0: Command failed: status = 0xc7,
flags = 0x51, unit #0.
Apr  7 21:27:58 allo last message repeated 3 times
Apr  7 21:27:58 allo kernel: SCSI disk error : host 0 channel 0 id 0 lun 0
return code = 8000002
Apr  7 21:27:58 allo kernel: Info fld=0x0, Current sd08:0f: sense key Aborted
Command
Apr  7 21:27:58 allo kernel:  I/O error: dev 08:0f, sector 544272
Apr  7 21:28:01 allo kernel: 3w-xxxx: scsi0: Command failed: status = 0xc7,
flags = 0x51, unit #0.
Apr  7 21:28:16 allo last message repeated 4 times
Apr  7 21:28:16 allo kernel: SCSI disk error : host 0 channel 0 id 0 lun 0
return code = 8000002
Apr  7 21:28:16 allo kernel: Info fld=0x0, Current sd08:0f: sense key Aborted
Command
Apr  7 21:28:16 allo kernel:  I/O error: dev 08:0f, sector 544280
Apr  7 21:28:20 allo kernel: 3w-xxxx: scsi0: Command failed: status = 0xc7,
flags = 0x51, unit #0.
Apr  7 21:28:35 allo last message repeated 4 times
Apr  7 21:28:35 allo kernel: SCSI disk error : host 0 channel 0 id 0 lun 0
return code = 8000002
Apr  7 21:28:35 allo kernel: Info fld=0x0, Current sd08:0f: sense key Aborted
Command
Apr  7 21:28:35 allo kernel:  I/O error: dev 08:0f, sector 544288


The tar command itself reported no errors.  Neither are there are ext3 fs
errors in the log.  But the disk errors seem to imply to me that there is
possible file corruption.  Is there anyway to identify which files are
sitting on the above sectors of the disk?

-- 
---------------------------------------------------------------
Paul Raines                   email: raines@nmr.mgh.harvard.edu
MGH/MIT/HMS Athinoula A. Martinos Center for Biomedical Imaging
149 (2301) 13th Street        tel:(617)-724-2369
Charlestown, MA 02129	USA   fax:(617)-726-7422

Hi,

On Tue, 2003-04-08 at 14:37, Paul Raines wrote:
> When tar'ing up an ext3 filesystem on a 3ware RAID I have, I got
> the following errors in my system log:
> 
> Apr  7 21:27:46 allo 3w-xxxx[957]: Drive error encountered on port 3 on
controller ID:0. Check cables and drives for media errors. (0xa)
> Apr  7 21:27:46 allo kernel: 3w-xxxx: scsi0: Command failed: status = 0xc7,
flags = 0x51, unit #0.
> Apr  7 21:27:58 allo last message repeated 3 times
...
> The tar command itself reported no errors.  Neither are there are ext3 fs
> errors in the log.  But the disk errors seem to imply to me that there is
> possible file corruption.  
Yep --- you are getting low-level SCSI driver errors, so chances are
that some of the writes are not completing correctly.
> Is there anyway to identify which files are
> sitting on the above sectors of the disk?
See the "icheck" and "ncheck" commands in debugfs.

Cheers,
 Stephen